Geometric orthogonal codes and geometrical difference packings (Q2161425)

Geometric orthogonal codes have been recently introduced because of their application in the synthesis of DNA nanomaterials. From a practical point of view, it is desirable to have a geometric orthogonal code with large number of codewords. For fixed \(n_1, n_2, w\) and \(\lambda\), the quantity \(\Phi(n_1\times n_2,w,\lambda)\) denotes the largest possible size among all \((n_1\times n_2,w,\lambda)\)-geometric orthogonal codes. \((n_1\times n_2,w,\lambda)\)-codes with size equal to \(\Phi(n_1\times n_2,w,\lambda)\) are called \textit{optimal}. In this paper, the authors provide upper bounds \(\Phi(n_1\times n_2,w,1)\) and, using the equivalence between \((n_1 \times n_2, w, 1)\)-geometric orthogonal codes and \((n_1 \times n_2, w, 1)\)-geometrical difference packings, different recursive constructions of optimal codes are provided. As a consequence, the exact value \(\Psi(n_1 \times n_2, 3, 1)\) is determined for any positive integers \(n_1\) and \(n_2\).